Comparison of Bayesian Spatiotemporal Models for Small-Area Life Expectancy: A Simulation Study

Abstract

Abstract The purpose of this study was to assess the precision, uncertainty, and normality of small-area life expectancy estimates calculated using Bayesian spatiotemporal models. We hypothesized 6 scenarios in which all 247 districts of South Korea had the same year-specific female population of 500, 1,000, 2,000, 5,000, 10,000, and 25,000 individuals during the study period (2013–2017). We generated 1,000 hypothetical data sets for each scenario and calculated district-year life expectancies. The precision and uncertainty of life expectancy estimates were compared between 2 Bayesian spatiotemporal models and the traditional method and Bayesian spatial models. We examined the normality of the life expectancy distributions generated by each method and investigated an optimal cutoff value for the comparisons. The Bayesian spatiotemporal models produced precise life expectancy estimates. However, the 95% uncertainty interval contained the true value with a probability of less than 95%. The Bayesian spatiotemporal models violated the normality assumption in scenarios with small population sizes. Therefore, life expectancy comparisons should be conducted using a cutoff value that minimizes false-positive and false-negative rates. We propose 0.8 as a cutoff value for determining the statistical significance of the difference in life expectancy.